DESCRIPTION: The hypotheses to be tested in this revised proposal is that loss of nuclear retinoic acid receptor-beta (RAR-beta) may play an important role in esophageal carcinogenesis and that restoration of this receptor may enhance retinoid actions on HEC cells. The hypothesis will be tested by using both cultured HEC cells and tissue specimens. There are four specific aims: 1. To analyze the expression of RAR-beta2 and RXR-alpha in premalignant and malignant tissues from Barrett's esophagus and HEC patients and to find out whether there is an association between RAR-beta expression and the expression of COUP-TFs, Nur77, the cellular RA contents, and differentiation/ proliferation markers (cytokeratins, involucrin, cornifin, and Ki67) by using in situ hybridization with digoxigenin-labeled antisense riboprobes and immunohistochemistry. 2. To determine the expression of COUP-TFs and Nur77 in cultured HEC cells by Northern blot and Western blot and to investigate their relationship to RAR-beta expression by transcriptional activation assay, gel retardation assay, and immunoprecipitation followed by Western blotting. 3. To determine the roles of RAR-beta2, RAR-beta4 in regulating cell proliferation, differentiating, or apoptosis in HEC cells. Retroviral vectors containing sense or antisense RAR-beta2 or RAR-beta4 will be stably transinfected into HEC cells to overexpress the receptors in cells lacking constitutive receptor or block the expression of the endogenous receptor in cells expressing constitutive receptor. 4. To determine the effects of certain receptor-selective retinoids on the expression of RAR-beta2 and on proliferation, apoptosis, and differentiation markers (i.e., cytokeratin 1, involucrin, and cornifin, type I transglutaminase) in HEC cells.